This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

Perspectives on James Robinson's article
are available by clicking the author's name below:
Robert M. Crane and Laura A. Tollen • Jay Gellert

H E A L T H  I N S U R A N C E : C O S T  S H A R I N G
W E B  E X C L U S I V E  

20 March 2002

Renewed Emphasis On Consumer Cost Sharing
In Health Insurance Benefit Design

Employers are demanding and insurers are offering insurance products with
higher deductibles and copayments, to offset premium increases.

by James C. Robinson


ABSTRACT:

Purchasers and health plans are reemphasizing deductibles, coinsurance, and other consumer incentives in response to renewed inflation and the continuing backlash against managed care. This paper explores the partial convergence of cost sharing and benefit design for preferred provider and health maintenance products and highlights experiments that foster price-conscious choice among benefit configurations, provider networks, systems of care, drugs, medical devices, and clinicians. Health insurance is evolving from comprehensive coverage for a restricted set of choices to limited coverage for a broader set of choices. Diverse benefit designs and increased consumer cost sharing challenge conventional policy wisdom but may counteract some of the pernicious features of the health care status quo.


A re-ignition of medical inflation and the continuing backlash against managed care are stimulating interest among purchasers and health plans in cost-control mechanisms for consumers, including thinner benefits, increased copayments, and higher deductibles. In the short term, such actions will extend the life of the health care status quo, allowing insurers to retain their existing portfolio of products, providers to retain their accustomed patterns of practice, and employers to retain their paternalistic approach to purchasing. In the long term, however, a sustained increase in consumer cost sharing will dramatically affect both sides of the health care market. Purchasers will move away from a one-size-fits-all approach toward more limited subsidies to support a broader variety of options. Plans will move from product uniformity toward multiple benefit configurations, discounted supplementary services, provider networks, systems of care, drugs, medical devices, and clinicians.

Benefit exclusions and consumer cost sharing exert only modest direct influences over the use, price, and quality of health care services, as many financial expenditures and clinical challenges occur for catastrophic or chronic conditions that exceed out-of-pocket payment limits.¹ One feature of the contemporary experimentation with benefit redesign, and what will distinguish it from the cost-sharing components of the bygone indemnity insurance era, is an attempt to leverage financial incentives directed at routine, low-cost services into incentives for cost-conscious choice and efficient delivery of catastrophic and chronic care. The most immediate manifestation of the trend toward health care consumerism is the raising of preferred provider organization (PPO) deductibles and health maintenance organization (HMO) copayments, the splicing of PPO features into HMO products through hospital admission coinsurance, the splicing of HMO features into PPO products through office visit copayments, and the addition of coinsurance to copayments in three-tier pharmacy benefits. The potentially more important manifestations of the trend include experiments with the medical savings account (MSA) product that permit higher deductibles and a broader set of (partially) covered benefits; variations in premiums and copayments according to the consumer’s choice of provider networks or delivery organizations; price differentials for use of centers of excellence for transplantation and networks-within-networks for chronic disease care; episode-of-illness allowances to motivate price-conscious shopping after diagnosis; and discount programs for products and services that fall outside insured benefits.

This paper explores the trend toward benefit redesign and increased consumer cost sharing in health insurance. It begins with the implications of increased cost sharing for PPOs and HMOs and then highlights experiments that would change health insurance from comprehensive coverage for a restricted set of choices to partial subsidy for a broad set of choices. The conclusion examines the policy challenges raised by new health insurance designs. Thin and diverse benefits foster risk selection, add to the complexity of the health care system, undermine cross-subsidies from the healthy to the sick, and create financial barriers to access for appropriate (as well as inappropriate) care. But they also counteract the tendency of open-ended tax subsidies to foster excessively rich benefits, facilitate a matching of product characteristics with consumer preferences, limit the effect of benefit mandates sponsored by special-interest groups, and promote a social perspective according to which health care is a scarce resource in need of priorities rather than an unlimited entitlement for which someone else can be forced to pay.

Cost Sharing In Managed Care Products

Benefit exclusions and cost-sharing provisions can be inserted into every form of health insurance product and thereby relieve some of the pressure on the monthly premium. The trend toward greater consumer responsibility for the cost of care is not neutral among alternative product designs, however. Benefit re-design is now contributing to the rising popularity of the PPO relative to the HMO product. While the PPO traditionally has been viewed as a quasi-indemnity product that offered broad choice but high cost, it lends itself more easily than the HMO does to the insertion of deductibles and coinsurance, which are much more effective than copayments are in offsetting premium increases. It remains to be seen whether continued cost inflation will stimulate the insertion of deductibles and coinsurance into HMO products to the degree already evident in their PPO competitors, in which case the HMO could retain its position as having the lowest premium in the insurance market.

Cost sharing in the PPO product. The typical PPO more closely resembles traditional indemnity insurance than the HMO does in terms of network design, provider payment method, utilization management, compatibility with self-insurance, and regulatory oversight and hence has grown primarily by drawing enrollees away from indemnity products rather than away from HMOs. The PPO contracts with a broad network of physicians and hospitals, partially reimburses claims from nonnetwork providers, pays mostly on a fee-for-service (FFS) basis, requires prior authorization only for hospital admission and high-cost outpatient procedures (or not at all), does not require primary care referral for specialty consultation, is easily adaptable to self-insured health benefit programs, and is exempt from the most onerous forms of state regulation and quality accreditation.² It once was conceptualized as “managed care lite,” a transition product on the road to the real thing, defined as an HMO with a narrow provider network, limited or no out-of-network coverage, risk-based provider payment such as capitation, extensive primary care gatekeeping and prior authorization requirements, and focus on fully insured business and subject to detailed regulation and accreditation. As the nation moves from a period of economic prosperity and premium stability to one of economic recession and cost inflation, however, the PPO’s distinctive feature turns out to be its similarity to the indemnity product in terms of consumer cost-sharing structures.

Copayments. Conventional indemnity insurance imposed a deductible, which makes the enrollee responsible for all costs up to a defined threshold, and then a coinsurance rate, which makes the enrollee responsible for a percentage of costs between the deductible threshold and an annual out-of-pocket maximum. As insurers shifted their indemnity enrollment to PPO products to compete with HMOs, they often restructured the cost sharing from coinsurance to fixed-dollar copayments for office visits, hospital admissions, and other services. Copayments offer enrollees the advantage of specifying in advance the amount for which they are responsible; they offer providers the advantage of ease of collection (co-payments typically are paid prior to service, while coinsurance typically is paid afterward). The copayment structure has been especially popular in the group (as distinct from the individual) market, since it offers visible financial support on the very first visit rather than only after a deductible has been paid. Copayments suffer from the salient limitation, as a cost-control instrument, of not varying according to which physician or hospital is chosen, despite the often considerable differences among providers in rates charged to the insurance plan. In contrast, percentage coinsurance provisions expose enrollees to at least part of the financial consequence of their choice. The economic prosperity and tight labor market of recent years protected consumers from substantial increases in premium contributions and cost sharing. The new benefit designs impose much more serious cost sharing, and enrollment appears to be shifting in the individual and small-group markets as the economy goes into a nosedive and medical inflation spikes, although it is too early to ascertain trends with clarity. Recent employer surveys and focus groups highlight an intention to dramatically increase consumer cost sharing in 2002 and 2003, while the available data from 2001 reveal only modest increases relative to previous years.³

Mix-and-match benefit design. Annual deductibles; special deductibles for brand-name drugs; copayment and coinsurance features for generic, formulary, and nonformulary drugs; office visit copayments; hospital admission deductibles and coinsurance; and special copayments for outpatient diagnostic tests, surgical procedures, and ancillary services can be mixed and matched to achieve whichever premium price point is desired by the purchaser. Benefit variability is almost infinite in the large-group market, as self-insured corporations specify an idiosyncratic mixture and then put it out to bid to multiple carriers. In the individual and small-group markets, however, health plans must balance the virtues of a broad product portfolio in matching prices with purchasers’ willingness to pay, on the one hand, with the virtues of a narrow portfolio in simplifying the subsequent product marketing and administration, on the other. Benefit designs in the corporate sector typically cover a broad range of services with little employee cost sharing, while those in the small-group sector offer thinner benefits at increased cost sharing, and those in the individual market cover the fewest services at the highest cost sharing. This variance in benefit design reflects the variance across sectors in administrative costs and concerns for adverse selection as well as the all-too-human tendency to spend other people’s money with less care than one’s own. In the corporate sector the purchaser and the consumer are strangers to one another, in the small-group sector they are personally acquainted, and in the individual market they are one and the same. It is more apparent in small than in large firms who pays the price of moral hazard, as high insurance contributions support comprehensive benefit designs, comprehensive benefits induce cost-unconscious choice of drugs and doctors, and cost-unconscious choice necessitates subsequent increases in premium contributions. Smaller purchasers suffer less than larger ones do from the tragedy of the insurance commons, where each subgroup of beneficiaries lobbies strenuously for coverage of its favorite service while acquiescing passively to the coverage of those favored by others.

Cost sharing and premiums. Insurers now commonly offer one or two HMO products plus a substantial number of PPO or, depending on the region, indemnity or point-of-service (POS) products, each with a different cost-sharing structure. Blue Shield of California, for example, offers two HMO and five PPO products in the state’s individual market (one HMO, one POs plan, and six PPOs in the small-group market), while Highmark Blue Cross Blue Shield offers one HMO, six POs plans, one PPO, and seven indemnity products in Pittsburgh’s individual market. Cost sharing can drive PPO premiums below HMO rates, even in highly capitated environments such as California. For example, Blue Shield’s average monthly premium for the zero-deductible, $10 copayment HMO is $289, approximately equal to the zero-deductible, $25 copayment PPO ($302) but much higher than the zero-deductible, $45 copayment PPO ($165), to say nothing of the $2,000 deductible, $45 copayment PPO ($120). Blue Cross of California brings the premium for a somewhat similar PPO product with a $2,500 deductible (and 25 percent coinsurance) down to $90 and a PPO with $1,000 deductible and no coverage for physician office visits down to $46.⁴

Cost sharing in the HMO product. In comparison to the dominant indemnity products, HMOs led the way in expanding benefit coverage to maternity, mental health, preventive, and pharmaceutical services and in reducing costs at the point of care by eliminating deductibles and replacing coinsurance with modest copayments. The cost-control features of the HMO were obtained almost exclusively from incentives focused on the supply, rather than the demand, side of the health care market. Close financial and organizational linkages with a cohesive provider network, prospective rather than FFS payment, primary care referral prior to specialty consultation, and various prior or concurrent review mechanisms limited use of high-cost procedures without asking the enrollees to bring out their checkbooks. But a decade of economic prosperity weakened the grassroots constituency for health care cost control and reinvigorated the demand for unimpeded choice of providers and procedures. Recent years have witnessed the step-by-step dismantling of the hallmark HMO product characteristics, as networks have been broadened to include most hospitals and physicians, payment has reverted from global capitation to partial capitation and FFS, and prior authorization and gatekeeping requirements have been relaxed or removed.⁵

Health plans now are striving to insert consumer cost-sharing mechanisms into their HMO products, to compete on price with the PPO. The easiest approach is to increase the traditional office visit copayment from $5 to $10, $15, and beyond. Even substantial increases in the traditional range of copayments do not exert a strong influence on the premium, however, as they constitute a small portion of spending on hospital inpatient, hospital outpatient, and pharmaceutical services. Some HMOs (such as PacifiCare, Health Net, and WellPoint) now can be purchased with more substantial cost sharing for hospitalization, such as $1,000 or $1,500 per admission copayments and additional 20 percent daily-charge coinsurance.⁶ Three- and four-tier pharmacy benefits insert higher copayment requirements for formulary but nongeneric drugs and, of potentially more importance, substitute percentage coinsurance for fixed-dollar copayment provisions for nonformulary (and hence very high cost) products. Analogous principles can be applied, although with greater administrative difficulty, to physician and hospital networks. Aetna’s triple-option USAccess product, for example, charges a modest copayment for services obtained with a referral and within the first-tier (HMO) network, a higher copayment for in-network self-referred services, and substantial coinsurance and deductibles for self-referred services obtained from noncontracted providers. PacifiCare has introduced an HMO product with differential copayments depending on whether the enrollee chooses a hospital that charges the insurer a high or low rate. POs products typically impose modest fixed-dollar copayments for services within the contracted network but a deductible and percentage coinsurance for services outside the network.

Effect on premiums. The effect of various permutations of deductibles, co-payments, and coinsurance on the premium is readily evident in a small-group market subject to guaranteed-issue regulation (which limits the influence of adverse selection on variations in product prices).⁷ In San Francisco a firm with four employees and four covered dependents could find coverage for as low as $531 using the WellPoint PPO with a $1,000 deductible, no coverage for office visits, additional copayments and coinsurance for hospital admission, and 20 percent coinsurance for ancillary services. The premium rises to $694 for a Health Net PPO with a $500 deductible, $20 office visit copayment (limited to two visits per year), additional copayment and coinsurance for hospital admission, and 20 percent coinsurance for ancillary services. A more conventional Blue Shield PPO with $1,000 deductible, $35 office visit copayment, additional copayments and coinsurance for hospital admission, and 20 percent coinsurance for ancillary services would cost $1,353, while a rich PPO product with zero (in-network) deductible, $15 office visit copayment, and 20 percent coinsurance for other services would demand $2,017. HMO products could be obtained for $1,136 from Kaiser Permanente (no deductible or coinsurance, $20 office visit copayment, no coverage for services outside the exclusive Permanente provider network), for $1,431 from WellPoint ($1,500 deductible, no coinsurance, $10 office visit copayment), for $1,498 from PacifiCare (no deductible, $20 office visit copayment, 20 percent coinsurance for hospital services), or for $1,815 from WellPoint (no deductible or coinsurance, $10 office visit copayment).

HMOs versus PPOs. It is unclear whether the loosening of supply-side incentives and strengthening of demand-side incentives will permit the HMO product to resume what once appeared to be an inexorable rise to dominance. After decades of continual growth, enrollment in HMOs leveled off and then began to decline in the past several years. Between 1996 and 2001 HMO enrollment declined from 31 percent to 23 percent of the employment-based insurance market, while PPO enrollment grew from 28 percent to 48 percent.⁸ The HMO is hampered by relentless bad publicity, ever-growing regulatory oversight and benefit mandates that often do not touch PPO products, administrative and legal complexities in being applied to self-insured health benefit programs, a local rather than national network structure, and, depending on the market, a dependency on fragile medical groups and hospital-based integrated delivery systems. Many multiproduct health plans that once emphasized their HMO products now are allowing or encouraging HMO members to roll their enrollment over into PPO products (for example, Aetna and Humana), while some health plans that once defined themselves as HMOs are launching or greatly expanding PPO products so as not to be left behind (for example, PacifiCare and Health Net). Blue Cross and Blue Shield plans, many of which lagged in shifting enrollment to HMO products, have enjoyed surging enrollment in their PPO products.⁹ Health plans most tightly linked to the HMO heritage increasingly conceptualize themselves in terms of health care delivery systems rather than as financing mechanisms and are withdrawing from markets where they lack strong provider relationships (for example, Kaiser Permanente and hospital-based HMOs). Many hospital systems that created HMOs are selling their products to conventional carriers, closing them altogether, or, in the case of the Providence health plan in Oregon and Washington, rolling their entire enrollment from HMO to PPO in one fell swoop.¹⁰

Experiments In Benefit Design

The shift from supply-side to demand-side incentives reflects the changing self-understanding of the health insurance industry. For two decades much of the industry embraced the principles of managed care and interpreted its role as one of reforming the health care delivery system through close affiliations with selected physicians and hospitals. But the financial turmoil among integrated delivery systems and the regulatory backlash against managed care have radically changed the industry’s strategy and vision. Rather than seeking to control costs by limiting consumers’ choice of providers, procedures, and products, health plans increasingly interpret their role as one of packaging health care services, pricing them at actuarially sustainable rates, gathering and disseminating information, promoting electronic connectivity among all participants, and otherwise getting out from between the consumer and the services the consumer wants to consume.¹¹ Incumbent health plans are designing new products and inserting new features into old products. Entrepreneurs are seeking to grow ideas into start-up firms that will both cooperate and compete with the incumbent plans. Most of the experiments will perish quietly, some will survive and remain restricted to narrow customer niches, but a few will flourish and propagate themselves throughout the health insurance ecosystem. While many variants are to be observed in the contemporary turmoil, the general tendency is for the new models to offer broader but more shallow coverage, more choices but at a lower subsidy, than the managed care mainstream. Three dimensions are worthy of attention: breadth of benefit coverage, choice of provider networks and organizations, and ease of access to individual clinicians.¹²

Broader choice of benefits. In the absence of consumer cost sharing, the addition of a new product or service to the benefit package exerts a strong influence on the health insurance premium. New coverage shifts the cost of previously used but noncovered services onto the insurer, and thence to the purchaser, and stimulates an increase in utilization as the out-of-pocket cost to the patient declines. When costs are stable and moral hazard is limited, employers like to accommodate employees’ requests for broader coverage, obtaining credit for benevolence while partially shifting the cost to the taxpayer through forgone tax collections and to the employee through forgone wage increases. But when costs are rising and moral hazard fears are strong, as in the case of complementary medicine, lifestyle drugs, cosmetic procedures, mental health, dental and vision services, and long-term care, employers find themselves resisting expansions of the benefit package.

The insertion of sizable copayments and deductibles alters the calculus of costs and benefits for expanding benefit coverage. Copayments targeted at particular services partially protect nonusers from the costs incurred by users, while allowing users to obtain volume price discounts and a partial insurance subsidy. Large deductibles relieve the insurer and employer of any expenditures for those benefits used by enrollees with annual expenses below the cost-sharing threshold. In the extreme, new services can be attached to the benefit package as noninsured but discounted options, thereby avoiding any pain for sponsors and non-users while allowing users to obtain lower unit prices from participating clinicians, pharmacies, and facilities. Aetna, for example, offers discounted but noninsured fitness services, vision services, and complementary medicine products. These services are self-managed, without referral or prior authorization, since they are paid directly by the enrollee to the provider at a rate schedule negotiated by the insurer (discounts are available only to enrollees in one of the insurer’s plans).

Cost sharing creates financial disincentives for the use of cost-effective and clinically effective services as well as for their more discretionary fellow travelers. Some health plans are exempting particularly valuable services, such as preventive care, from deductibles and copayments altogether, while varying the cost-sharing requirements for other services. Office visit copayments can be higher for specialty than for primary care and higher for patient-initiated specialty visits than for those obtained upon referral. Emergency room visits generate substantial copayments that may be waived if the patient is admitted to the hospital (a rough-and-ready indicator of whether the event was a true emergency or an after-hours primary care visit). Copayments for inpatient care can exceed those for outpatient testing, surgery, and rehabilitation.

Medical savings accounts. One way for employers and insurers to expand the potential range of benefit coverage without raising the premium is through variants on the MSA product. The MSA has become a political football, eulogized by conservatives as the one-size-fits-all solution to health system woes and vilified by liberals as an invitation to risk selection. In practice, the MSA is a benefit design that allows sponsors to use pretax dollars to fund a savings account out of which the beneficiary can pay the costs of health care underneath the deductible; an indemnity or PPO wraparound product is added on top of the deductible to cover catastrophic expenses. The presence of the MSA permits the deductible to be raised far above the levels observed otherwise, with cutoffs of $3,000, $5,000, and $10,000 already in the market.

Variants on the MSA product are being promoted by major health plans, often in conjunction with PPO network designs. Aetna U.S. Healthcare, long the champion of narrow-network HMO products, has created a nationwide PPO product with a high deductible and employer-funded savings accounts.¹³ The MSA design is being promoted most enthusiastically, however, by startup firms such as Definity and Lumenos, which combine Internet-based information-dissemination and decision-support tools with MSAs and a PPO-based wraparound design. These products are marketed to corporate employers, which contribute a dollar amount that is split between funding the high-deductible PPO premium and funding the savings account. Unspent MSA balances can be rolled over from year to year, so a family with uneven expenses may never use its own money (but only the employer’s contribution) while continually making trade-offs between spending and saving. Unspent balances may be forfeited if the employee leaves or loses the job. The employer can specify a very broad set of services to be included in the benefit design, since the incremental cost of using a particular service is borne fully by the employee in terms of a reduced MSA balance available to fund other services. Preventive services are fully covered without drawing on the MSA.¹⁴

Broader choice of provider networks and systems. The original principle of managed care was to rely on physicians who combine a culture of cooperation, administrative efficiency, and financial joint destiny through prepaid group practice or physician-hospital integration. In the absence of well-functioning provider organizations, however, most HMO products came to rely on networks of individual and small-group practices that accepted lower fees and some clinical oversight in exchange for a higher volume of patients. To achieve these volume discounts, health plans contracted with only a limited number of physicians and hospitals in each community, thereby forcing some enrollees to switch providers to use their insurance coverage.

An emerging set of health insurance benefit designs seeks to retain some of the advantages of provider coordination while broadening consumer choice. Rather than arm-wrestling with doctors and medical groups under the implicit threat of network exclusion, these insurance products include any willing physician and provider organization but pass the differences in fee levels on to the consumer through higher premiums or copayments. At the extreme, these insurance product designs do not negotiate fees at all, creating a market that permits providers to charge whatever they think their patients are willing to pay and that permits consumers to choose among all providers rather than be limited to a contracted subset. The premium charged to the employer covers most or all of the fees charged by low-cost providers, while the employee pays the full incremental cost of the fees charged by more expensive providers.

The most prominent instance of cost-conscious choice of provider organizations under the umbrella of a broad-network benefit design has been the self-insured PPO pioneered by the Buyers Health Care Action Group (BHCAG) in Minneapolis and subsequently spun off into a startup called Patient Choice Healthcare.¹⁵ BHCAG develops networks of care systems, which include multi-specialty clinics, independent practice associations (IPAs), and hospital-centered entities, that set their own FFS payment schedules and manage their own practice patterns. The health plan uses claims experience, risk-adjustment methods, and actuarial models to generate premiums based on these fee schedules and passes these premiums through to purchasers. Enrollees who choose care systems that charge high fees must pay a higher premium than those who choose a low-cost system. Employers establish a defined contribution no higher than the lowest care-system premium, thereby requiring the employee to pay the full incremental difference if an expensive care system is chosen.

Insurance benefit designs that feature cost-conscious choice among medical groups face two salient dilemmas. Most US communities lack care systems that possess the clinical and administrative capabilities to provide the full range of services; health plans hence must cobble together networks of disconnected primary care, specialty, and subspecialty physicians. Even where medical groups, IPAs, and physician-hospital systems are present, some patients want to make their own selection of particular specialists instead of going along with the affiliations that the physicians have made among themselves. Some new benefit structures create incentives for consumers to customize their own provider networks. These can either supply the rudiments of coordination in an otherwise fragmented medical community or permit consumers to modify the components of existing provider-developed care systems. Cost-sharing provisions expose consumers to the economic consequences of the fee and practice patterns of the particular clinicians they select.

One ambitious model has been developed by Vivius, a startup working with established health plans in several markets, in which each enrollee would select one physician in primary care and in each of the principal specialties, a set of facilities (for example, hospital and ambulatory surgery center), and various ancillary services.¹⁶ It is assumed that most enrollees will select the specialists, facilities, and ancillary providers suggested by the enrollee’s preferred primary care physician. Where physicians already practice in multispecialty medical groups, consumers typically will select the group for all of their needs but retain the right to selectively add and subtract particular providers. Each Vivius enrollee customizes a care system and pays a premium aggregated from the fees charged by each physician, facility, and ancillary provider within the system. Providers determine their own fees on either a capitated or FFS basis, the cost being passed to the consumer through differential premiums. A high-deductible indemnity wraparound covers the services of subspecialists, out-of-area utilization, and other items not provided by the core self-defined network. Decision-support software permits enrollees to prospectively evaluate the cost implications of substituting a new physician or facility into the customized network or of going outside the network and filing claims to the insurance wraparound. This structure exposes the consumer to the cost implications of his or her choice of providers more fully than conventional coinsurance does, since the monthly premium for each customized network captures 100 percent rather than merely 20 percent (the typical coinsurance rate) of the differences among providers in utilization patterns and unit prices.

Broader choice of individual clinicians. Many patients want unfettered choice of specialists and facilities in the event of a change in health status, despite having previously chosen an integrated delivery system. For these consumers, the unit of choice needs to be the individual physician rather than (or in addition to) the provider system. POs products, which offer comprehensive coverage for the use of contracted physicians but higher consumer cost sharing for the use of out-of-network physicians, offer a partial solution to this dilemma. However, the administrative complexities of navigating multitier products and the increasingly successful provider pushback against contractual discounts are reviving an old technique for inducing patients to select low-price providers. Classical indemnity coverage allowed providers to bill patients at whatever price was mutually agreeable and then specified a schedule of payments the insurer would make to (partially) reimburse the patient. Patients who chose high-price providers thus paid 100 percent of the incremental fee above the indemnity schedule. This contrasts with subsequently popular service-benefit insurance (pioneered by Blue Cross) and managed care products, which paid providers directly at a contracted rate and left the enrollee responsible for only a portion of the incremental fee (through coinsurance) or shielded the patient altogether (through copayments that did not vary across providers).

Internet health plans. The emergence of Internet-based business-to-consumer and business-to-business marketplaces, replete with auction, reverse-auction, and menu price mechanisms, has stimulated interest in analogous markets for medical care services. As in other Internet markets, the guiding principle is open participation by all willing buyers and sellers (no network exclusions) and mutual agreement on prices by buyers and sellers (no network contracts and hence no network discounting). HealthAllies, HealthMarket, and other start-ups have established business-to-consumer Internet markets of this nature and have had some success for services outside the covered benefits. HealthMarket has been most aggressive in seeking to combine principles of indemnity payment, any willing provider, and price transparency into an insurance product that eschews network restrictions while motivating price-conscious consumer choice.¹⁷

HealthMarket conceptualizes the key unit of care and hence of payment as the episode of illness, rather than either the individual procedure (paid through FFS) or the full continuum of care (paid through capitation). The episode of illness would also be the foundation for the structure of insurance benefits received by the enrollee, using indemnity principles to create a severity-adjusted budget for each condition or major procedure and allowing the consumer to shop the entire range of providers to obtain the blend of price and quality that best matches his or her preferences and willingness to pay. In the HealthMarket product, consumers can choose from among all networks that offer care for a particular type of illness episode after, rather than before, being diagnosed. For example, patients with cancer can select among oncology networks (after diagnosis) rather than being forced to accept whichever oncologists happen to be included in the medical group they chose at open enrollment (before diagnosis).

HealthMarket is developing a series of indemnity-style products, each with a greater reliance on episodes of care as the basis for enrollee benefits and provider payments, respectively. The first product is a conventional high-deductible PPO, with a flexible spending account to cover costs incurred below the deductible, an out-of-pocket annual maximum payment, and twelve procedure-specific episode allowances that collectively account for 10 percent of physician and hospital charges to an enrolled population. The initial set of allowances is designed around procedures for acute conditions, rather than chronic disease care, because it is easiest for consumers to shop and for providers to price such procedures.¹⁸ Claims incurred for services related to the initial twelve procedures are paid on a first-dollar basis rather than being subject to the deductible or coinsurance, but the total payment by the plan is limited to the prespecified, procedure-specific budget. These budgets function as medical debit and credit accounts. If a comorbidity or complication arises, the insurance plan credits the consumer’s budget with additional funds to cover the needed extra services; when the consumer incurs a service, the provider claims are debited from the account. The consumer is thus at financial risk not for the first dollar but for the last dollar of care (up to the out-of-pocket maximum), which implies that he or she pays the full difference in procedure prices charged by different providers but faces no financial disincentive to initiate care (for example, no deductible) and to shop for less costly providers.

The second HealthMarket product will include several hundred procedure and chronic condition allowances (for example, asthma, diabetes, and congestive heart failure) that collectively will account for half of the physician and hospital costs (the other half of costs will fall under the conventional high-deductible, flexible savings account, PPO product).

Challenges For Health Policy

The mainstream of health policy traditionally has evinced skepticism about, if not hostility to, multiple insurance benefit designs and consumer cost sharing. The virtues of uniform and comprehensive benefits have impressed the otherwise fractious advocates of single payer, managed competition, and hybrid proposals to solve the problems of the health care system. In light of the widespread backlash against both public-sector taxation and private-sector managed care, however, a reconsideration of the conventional criticisms is imperative. Each of the four standard objections, while retaining great force, must be balanced against potential advantages of deductibles and diversity in the design of health insurance benefits.

Risk selection. Consumer choice among diverse benefit designs inevitably raises concerns of adverse selection, as sicker patients choose more-comprehensive coverage in anticipation of frequent use, while their healthier neighbors select thinner benefits in anticipation of infrequent use. The premiums of each product come to reflect the illness profile of the purchasers as well as the actuarial cost of the designs themselves, eventually forcing comprehensive packages beyond the range of affordability.¹⁹ The abstract logic of adverse-selection models is unimpeachable.²⁰ The social cost inflicted by consumer choice among multiple benefit designs in the real world may be questioned, however.²¹ The US health care system seems afflicted more by excessively thick than by excessively thin benefit designs, contrary to what would be inferred from the discussion of risk selection. The uncapped tax subsidy for employer-based contributions, the paternalistic obfuscation of the fact that employees pay for insurance benefits themselves through forgone wage increases, and the multitude of governmentally mandated benefits combine to generate insurance designs that are more comprehensive than consumers would choose if they paid directly with after-tax dollars. It is possible, however, that the contemporary variation in benefit designs will induce more-aggressive risk selection than witnessed in the managed care era, when products differed by provider network more than by benefit structure.

Administrative complexity. Diversity in benefit coverage and cost-sharing requirements across clinical services and insurance products compounds the already horrific administrative waste in the US health care system. Patients find it difficult to understand their coverage and often face “sticker shock” after an episode of care. Physicians and hospitals find it administratively difficult to collect the coinsurance they are owed and must hire additional clerks and accountants to battle those hired by the insurance entities. The cognitive challenges and administrative costs imposed by diversity in benefit design are no different in kind, however, from those imposed by diversity in other arenas of social life. Life would be cheaper and simpler if everyone wore the same outfit, played the same sport, drove the same car, bathed with the same soap, and believed in the same God. We shun uniformity of clothes and creed to foster innovation and to celebrate the virtues of diversity and individuality. The most abstruse aspects of benefit diversity presumably could be alleviated through boundaries on the range of options while permitting consumers to match product features to personal preferences and change their minds every once in a while.

Cross-subsidies. Copayments and deductibles are not adjusted for enrollees’ income or health status and thus impose much greater burdens on the poor and sick than on the rich and healthy. Consumer cost sharing is a regressive method for financing health care and a corrosive threat to the principle that care for the unfortunate is the responsibility of the fortunate. Whatever the virtues of uniform and comprehensive benefits when supported by progressive financing techniques, however, the imposition of uniformity and comprehensiveness without such financing aggravates existing injustices. Forcing the poor to purchase comprehensive insurance while forgoing goods and services to which they attach higher value is not doing them any favor. Mandating inclusion of every provider, product, and procedure for which there exists a politically mobilized lobbying organization transfers resources from the weak to the strong.

Financial barriers to access. Deductibles, coinsurance, and copayments are designed to reduce use of those services that patients feel offer less value than they cost. It goes without saying that consumers and clinicians often hold different perspectives on the value of particular diagnostic and therapeutic interventions. While the RAND Health Insurance Experiment found no substantial health effects of coinsurance-related reductions in utilization, it is to be assumed that substantial cost sharing will lead to some adverse outcomes for some patients resulting from forgone care.²² The policy implications of this unfortunate fact are not obvious, however. The absence of meaningful cost-sharing components to the standard HMO product contributed enormously to the culture of entitlement and backlash against managed care techniques such as disease management, primary care focus, and integrated delivery systems, despite the potential of these mechanisms to improve quality as well as to reduce cost. The only way now to channel patients with severe conditions into narrow networks, thereby obtaining the economic and clinical benefits of scale and learning-curve effects, is to use carrots rather than sticks. Patients who reject mandated use of narrow networks, gatekeeping, and protocols may accept them in exchange for selective reduction in cost-sharing requirements.²³

One of the great truths of economics is that unsustainable trends will not be sustained. Double-digit inflation in health care spending within an economy that otherwise grows at single-digit levels is one of those unsustainable trends. Managed care, as conventionally defined to include narrow provider networks, global capitation, gatekeeping, and prior authorization, appears to be a second. A sustainable trend would include a rate of medical inflation modestly higher than the overall rate of economic growth in an ever older and more prosperous society, bolstered by culturally legitimate and economically effective mechanisms to allocate scarce health care resources among competing ends.

Consumer cost sharing may contribute to bottom-up health system reform after the exhaustion of governmental and corporate initiatives. In the short term, increased coinsurance and deductibles may cool the demand for physician visits, routine tests, and brand-name drugs, partially offsetting the surge in medical inflation. In the medium term, MSAs, service-specific copayments, episode-of-care pricing, three-tier pharmacy and network benefits, and other demand-side innovations may help to address the challenges raised by chronic care. In the long term, thin benefit designs may foster a grassroots constituency for affordability and hence for the use of technology assessment and cost-effectiveness analysis in health care. Americans’ first-best health care preference is cost-unconscious choice, with some distant, unknown party footing the bill. When faced with the second-best trade-off between cost-conscious choice and no choice at all, however, Americans may grumble but select the former.

This research was supported by Kaiser Permanente and the California HealthCare Foundation.

NOTES

1. M.L. Berk and A.C. Monheit, “The Concentration of Health Care Expenditures, Revisited,” Health Affairs (Mar/Apr 2001): 9–18.
2. See R. Hamer and D. Anderson, PPO Operations and Markets (Minneapolis: InterStudy, February 2000); and L. Greenberg, Perspectives on PPO Performance Measurement from Consumers, PPO Leaders, and Employers (New York: Commonwealth Fund, September 2000).
3. For 2002 and 2003 intentions, see L.C. Marsh et al., Lehman Brothers Healthcare Benefit Brokers Conference Call—He Says, She Says (New York: Lehman Brothers Global Equity Research, 15 August 2001); L.C. Marsh et al., Lehman Brothers 2001 Mid-Year Fortune 500 Managed Care Survey (New York: Lehman Brothers Global Equity Research, 26 July 2001); Watson Wyatt, Changing Role of Health Care Benefits, 2001 (Washington: Watson Wyatt Worldwide [jointly with Washington Business Group on Health and Healthcare Financial Management Association], 2001); M. Freudenheim, “Employees Are Shouldering More of Health Care Tab,” New York Times, 10 December 2001, C6; G.P. Mays, R.E. Hurley, and J.M. Grossman, Consumers Face Higher Costs as Health Plans Seek to Control Drug Spending, Issue Brief no. 45 (Washington: Center for Studying Health System Change, November 2001); and A. Howell, “Employers, Plans Use Short-Term Fixes such as Cost-Shifting to Address Concerns,” Health Plan and Provider Report (January 2002): 40–43. For 2001 data, see J. Gabel et al., “Job-Based Health Insurance in 2001: Inflation Hits Double Digits, Managed Care Retreats,” Health Affairs (Sep/Oct 2001): 180–186; and B.C. Strunk, P.B. Ginsburg, and J.R. Gabel, “Tracking Health Care Costs,” www.healthaffairs.org, 26 September 2001, W39–W50.
4. Blue Shield of California and Blue Cross of California are independent and competing insurance companies. Blue Cross of California is a trademark of WellPoint Health Networks. These rates apply to a single forty-four-year-old enrollee in the San Francisco region in October 2001.
5. D.A. Draper et al., “The Changing Face of Managed Care,” Health Affairs (Jan/Feb 2002): 11–23.
6. California HealthCare Foundation, “MCOs to Introduce Network-within-Network Plans Featuring Steep Copays for Certain Hospitals,” www.californiahealthline.org, 24 October 2001.
7. These price quotes were obtained 11 April 2001 from ehealthinsurance.com by Laura Tollen of the Kaiser Institute for Health Policy. Products differ in additional dimensions, but all include coverage for pharmaceutical products. L. Tollen, “Policy Issues in Health Benefit Design” (Paper prepared for the roundtable “Connecting Public Policy to Health Benefit Design,” Kaiser Institute for Health Policy, Oakland, California, September 2001).
8. Gabel et al., “Job-Based Health Insurance in 2001.”
9. R. Cunningham and D.B. Sherlock, “Bounceback: Blues Thrive as Markets Cool toward HMOs,” Health Affairs (Jan/Feb 2002): 24–38.
10. Technically, Providence switched to an exclusive provider organization (EPO) product, which is a PPO with no out-of-network coverage.
11. J.C. Robinson, “The End of Managed Care,” Journal of the American Medical Association 285, no. 20 (2001): 2622–2628.
12. The following discussion derives from the author’s case study of health insurance startups, including site visits and case studies with innovators, incumbent health plans, purchasers, suppliers, investors, and analysts. For additional discussion of these firms, see J.B. Christianson, S.T. Parente, and R. Taylor, “Defined-Contribution Health Insurance Products: Development and Prospects,” Health Affairs (Jan/Feb 2002): 49–64.
13. California HealthCare Foundation, “Aetna to Offer Combined PPO/Health Savings Plan in 2002,” www.californiahealthline.org, 20 September 2001.
14. Information on Definity products can be obtained from www.definityhealth.com.
15. J. Christianson et al., “Early Experience with a New Model of Employer Group Purchasing in Minnesota,” Health Affairs (Nov/Dec 1999): 100–114; DA Kindig, ed., Value Purchasers in Health Care: Seven Case Studies (New York: Milbank Memorial Fund, 2001); and J.B. Christianson and R. Feldman, “Evolution in the Buyers Health Care Action Group Purchasing Initiative,” Health Affairs (Jan/Feb 2002): 76–88.
16. For information on Vivius, visit www.vivius.com.
17. For information on HealthMarket, visit www.healthmarket.com. For a detailed analysis of the episode of care pricing model, see D. Emery, Global Fees for Episodes of Care: New Approaches to the Purchasing of Healthcare (New York: McGraw-Hill, 1999).
18. The twelve procedures include appendectomy, open and laproscopic cholescystectomy, hernia repair, hemorrhoidectomy, vaginal and cesaeran delivery, knee and shoulder arthroscopy, carpal tunnel surgery, and lithotripsy.
19. F.J. Hellinger, “Selection Bias in Health Maintenance Organizations: Analysis of Recent Evidence,” Health Care Financing Review 9, no. 2 (1987): 55–63; H.S. Luft and R.H. Miller. “Patient Selection in a Competitive Health Care System,” Health Affairs (Summer 1988): 97–119; and J.P. Newhouse, “Reimbursing Health Plans and Health Providers: Efficiency in Production versus Selection,” Journal of Economic Literature 34, (1996): 1236–1263.
20. G. Akerlof, “The Market for Lemons,” Quarterly Journal of Economics 84, no. 3 (1970): 488–500; and M. Rothchild and J. Stiglitz, “Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information,” Quarterly Journal of Economics 90, no. 4 (1976): 629–649.
21. M.V. Pauly, “Taxation, Health Insurance, and Market Failure in the Medical Economy,” Journal of Economic Literature 24, (1986): 629–675.
22. R.G. Brook et al., “Does Free Care Improve Adults’ Health?” New England Journal of Medicine 309, no.2 (1983): 1426–1434.
23. WellPoint, for example, will waive additional copayment and coinsurance requirements in multitier pharmacy benefit designs for chronically ill patients who agree to participate in disease management programs. Mays et al., Consumers Face Higher Costs as Health Plans Seek to Control Drug Spending.

Jamie Robinson is professor of health economics at the University of California, Berkeley.

©2002 Project HOPE–The People-to-People Health Foundation, Inc.